The Inside Track on Firefox Development.

« Where Did Firefox Come From? | Main | Camino 1.0 »

February 10, 2006

No Firefox2 a1 Today

Despite much reporting, and earlier schedules indicating that Firefox2 a1 would be released today, the reality is a little different. Read Mike Beltzner's note from the Firefox2 dev meeting on Tuesday. We're pushing it off a little so that we can actually have some new features for you to play with.

Dogfood

Dogfood is a term in computer jargon that means software has reached a point that it's basically usable, even if some functions are missing, and others are clumsy. It means it's something you can use as a day-to-day browser. The goal for Firefox2a1 is to reach dogfood on several of our key features.

Posted by ben at February 10, 2006 10:09 AM

Comments

Better late than rushed. :\

Posted by: ant at February 10, 2006 10:59 AM

As long as the branding for Branch (and Trunk) are sorted out before Fx2α1, I don't mind when it lands... :D

Posted by: Jonathan Stanley at February 10, 2006 11:52 AM

Will FF2a1 be usable as a "day-to-day browser", or will it be as buggy as FF1.1a1?

Posted by: Nanaki at February 10, 2006 12:44 PM

Everyone who downloaded Firefox 1.5b2 is on the beta channel. Dogfood releases really shouldn't be served up to that many people.

Posted by: Ben Basson at February 10, 2006 1:00 PM

@Ben Basson
There's a difference between Betas and Alphas ;)

@Ben
wasn't a delay expected? when I read about the plans for 2.0 the first time, I was very quickly sure that this deadlines would never be held in... Wouldn't a little bit more realistic roadmap be a good idea?
don't see this as a blame, it's just a suggestion... this could avoid a little bit miss-understanding and confusion among people that are interested in firefox but aren't following it that closely as the reader of planet mozilla do.

Posted by: Sebastian at February 10, 2006 1:22 PM

Do alpha releases get shipped to people on the beta channel?

Posted by: Ian at February 10, 2006 1:52 PM

Sebastian: IMO (and this is just mine, not anyone else's) setting "realistic" deadlines is both hard and not that useful. The more time you make it seem there is in a schedule, the more time people spend procrastinating or wasting time on non-priority items then rush to get the real work done by the deadline later, causing instability and bigger slippages. I'd rather we say we're going to ship earlier and slip as necessary.

Posted by: Ben at February 10, 2006 2:55 PM

Ben, when will the 1.8 branch get Bon Echo Alpha 1 branding and the version bump to 2.0a1? Mike Connor said in Bugzilla that artwork was only holding it back, and that was over a month ago.

Posted by: Rishi Maharaj at February 10, 2006 4:12 PM

@ Sebastian: That's exactly my point. See the meeting minutes where the issue of releasing a1 on the beta channel was raised. It is a terrible, terrible idea.

Posted by: Ben Basson at February 10, 2006 4:54 PM

@Ben Basson
oh sorry I missed that. doing so wouldn't be good at all, especially since there were many claims that people received the beta that wouldn't really have wanted to do so. ok, it's their own fault if thei already loaded the 1.5 beta, but giving them alpha versions wouldn't be good at all, i think.

@Ben
ok that's a good point. but if I look at the current release plan - most features that should be included in the trunk for alpha 1 aren't there, at least the most visible ones.
so i think something goes wrong with the release planning. probably some more strict rules what has to be done by the MoCo-developers would be a possibility, but ok, that's probably difficult, too.

nevertheless, you made firefox enormously successful, and you know what to do. at least you reacted to my post, what made me already very happy :)

Posted by: Sebastian at February 11, 2006 3:02 AM

Please DO pay attention.

Firefox 1.5 was boring.
It consisted on "under-the-hood" changes, at best. And bugfixes of things that should have been right the first time around, at worst.

A smart user does not necessarily see with bad eyes those not-so-visible changes.
But let me put this clearly, wide and loud: the changes in the current timelines for firefox 2.0, and even (gasp!) for firefox 3.0, are ALREADY boring!!!

And those are months and months away!!

This is, of course, unacceptable.

Luckily, there seems to be some consensus that changes in firefox 2.0 should be of the easily-notable-by-end-users, non-boring type.

However, and EVEN considering that, I think all is wrong already, anyway.

Why so? Plenty of reasons, keep reading...

-------------------------------------------

Problem number 1: Some extensions should obviously (by now) be included in the main code. And they are not getting there.
So we keep having to add extensions for features that should NOT be extensions. Heaps and more heaps of extensions.

Oh, yeah, yeah, don�t bother, I can hear you already:
"That is a feature, not a bug. You can make firefox as big or small as you want".
And...
"Not everyone would agree with which features should be incorporated and which should not."

Em...no, sorry.
There might be some not-so-clear cases. Sure.
But there are some clear ones. And there is not even mention of including them, as far as I can see today.

We can discuss extension by extension, or feature by feature. I can make some strong cases for some of them. But I will not here, now.
My point here is that not even the *idea* seems to be considered at this time. Hughe mistake.

-------------------------------

Problem number two: No killer features.

I do not give a peanut if some other icon now gets customizable.
They should have all been customizable, from the beginning. Systematically so. And even from before the beginning.
No, this is no "exciting feature".

Fancy to have a look at what would be some serious features?
Lets have a look at firefox today, from a user perspective...
Here we go...

We do not need two separate things, an adress box, and a search box. We can do with just one, period. Furthermore, it could also include a third function, namely search and highlight within the page.

Here I must go even much more radical: that unified bar, should not even be within firefox!!!!
Yes, you read correctly!!!!!!!!
Repeat with me

"In firefox, there should not be *any* adress box."

Now you think I�m crazy, dont ya?

Let me explain...

In the whole of my computer, I should have only ONE box.
That box should be accesible system-wide with just one key combination (windows+s works well for windows).
That box should not be in any browser, but in my desk bar (including being able to plug it into useful modifications, such as True Launch Bar).
That box should allow me to:
1) Search in my computer (including interface with indexing apps, such as desktop searchers).
2) Plug a URL.
3) Do a web search (including being able to choose between different engines).
4) Being able to search (and optionally highlight) in currently open documents (aka, search in page...but also perhaps search in *any* open document, even in other apps).
5) Optionally, ability to function as a command line as well.
(Yes, this last point is a bit risky in terms of security, I'm aware. But it could be done, And it should.)

That would be *some* seriously good thing. But there is more...

----------------------------------

We do not need bookmarks anymore.

Yes, you read correctly.
The whole lot of everything that says "bookmark" today, should go.

How come? Easy, just one word: scrapbook.
(I personally would prefer another name, but that extension called scrapbook today is what I have in mind).
If you have already included what the scrapbook extension does today, then a "bookmark" is just a particular case of a web capture (the one which stores just the address, and not a snippet content, nor the whole page, nor the whole site).
So goodbye special "bookmarks" interface items.
Furthermore, one does not even need to file captured content anymore either. One can simply capture, then search, gmail style.

------------------------------

Next, extending scrapbook functionality to the logical end:

What are the basic operations you do with web content?
1) View it
2) Store it
3) Edit it.
4) Repost it

One should be able to follow a link, and if interested, saving the desired content, be it the whole site, just a whole page, or some selected parts of it. Then one should be able to further edit that captured content. Then one should be able to publish it again (for instance, to a forum or blog).

That's the whole, clearly ordered information cycle.
Easily understandable.
And firefox should be able to do it all.

The web is not, and never was, intended to be one way. It is two ways.
If we wanted one way only, we had the idiot box already (TV).
So we need an editor, and some post-to-forum/blog thing.

---------------------------------

So what's the new, emerging picture?

We get a unique, system-wide box. For everything. In the deskbar. Accesible by a unique, simple system-wide key combination at any time, without needing to switch apps. Without even needing TO LOOK AT THE SCREEN.

In the browser interface itself, we get more space (valuable real state), and we get a system to capture, re-process and re-publish any content: the whole information cycle.

-----------------------------------

Finally, let me remind you where IE7/Vista might be going.
The key word is "winFS".
The whole file system becomes (like) a database.
Forget folders. You do not file, you search (see gmail).
Because of this, you can optionally tag (the scrapbook extension already allows for this).
Scrapbook, even today, works a little like a database.

So do not get left behind, I say.

-------------------------------------

And if you are thinking all this is too much work, or too radical, let me remind you that Opera already has voice operated browsing.

I hope I�m clear.

Fernando Franco

P.D: Regarding tags (metadata):

A WHOLE new level of functionality is coming. It is called the Semantic Web.
Tagging is perhaps its base.
Be ready for it.

Posted by: Fernando Franco at February 12, 2006 1:01 AM

Fernando that�s the problem that�s just your opinion of what�s constitutes a good browser (I'd like to see this consensus you speak of) personally speaking I find some of what you suggest horrific and I would stop using Firefox the moment it went in that direction.

The current system for the most part works because you simply cannot please everyone by including their pet extension you will turn Firefox in a bloated app, Firefox is in a fashion trying to convert IE users what you are proposing would bamboozle most IE users and make them turn away from Firefox.

Anyway Ben I hope the native rendering of menus under Windows makes it into 2.0 that would be a nice feature, its the little things that count : )

Posted by: Jake Evens at February 13, 2006 2:57 AM

It's the little things that count for user like us, on the bleeding edge. Such things make no difference to normal users: the users Firefox is left to win.

I'm not saying that these things shouldn't be done but Firefox is supposed to be different in that it's not aimed at just a techie audience.

For example, I think the everything should be an extension, not a core feature approach is harmful to these users. Extensions are *not* discoverable enough. You do not get ordinary users browsing addons.mozilla.org, and nor do I for that matter, but every odd time I do, I find a useful extension. Secondly, extensions do not experience the same code reviews as core features and so the Firefox team cannot rely on their resistance to bugs and conformance to UI guidelines and such.

Extensions are great for power users, but it is necessary to migrate useful features from extensions to the Firefox core build. For example, WinAmp is very popular among techheads, but now iTunes has stolen lots of market share. One reason for this I think is out-of-the-box utility rather than relying on extensions. Imagine if there was a PodCasting extension and a Shop extension.

Fernando's great vision is one I sympathies for, though it is not the place for Firefox to enact it. It would need to be part of the OS or at least the shell. Perhaps Linux would be a candidate for this sort of innovation since the code is open source. It wouldn't surprise me if this was something we see in Windows versions to come in the distant future.

Don't get me wrong, I love Firefox, but when I downloaded IE7B2 the otherday I loved that too; it forced me to remove my rose-tinted glasses. Firefox is great but I think it could do with a bit more innovation.

Posted by: Bruce Boughton at February 13, 2006 3:09 AM

Bruce - I actually think the success of iTunes is not related to it having more features, it probably has fewer overall than WinAmp (or at least it seems that way from a user's perspective) - it is related to the simple fact that when you start iTunes and put in a music CD you can hit ONE BUTTON to rip it into your library. Everything else within that app follows that elegantly simple model. In short: it just works.

Posted by: Ben Goodger at February 13, 2006 2:43 PM

Just please fix memory prob!

Posted by: dave at February 13, 2006 8:06 PM

First, I would like to say that I disagree with the bulk of what Fernando said. If he wants all that kind of stuff he should move to Opera, which is a power user's browser.

That said, he is right about a few things:
1. Firefox 1.5 is boring. Don't get me wrong it is a nice update, but there is just no "WOW" factor in it. Firefox 2 should have at least one new feature that turns heads. I know that the auto update feature was meant to that, but it isn't used enough and even Microsoft can do that much with its clunky apps. Also, the moveable tab feature is only a feature if you use tabs. I'm pretty sure that most people to this day don't know that Firefox even has a tab feature.
2. The bookmarks system needs to be fixed. I don't think that it should be taken as far as Fernando said, but I have had troubles trying to synch my bookmarks up between two computers that I use. It would be nice if the bookmarks were stored in a cleaner format.

Second, if Mozilla wants Firefox 2 to be successful , it is going to have to have the memory leak problem solved out of the box. I don't like the idea of having to crawl around the about:config set up on a gold release to fix a problem that should have been solved in the beta releases.

Third, I have downloaded IE7 Beta and tested it out a little bit. Right now I found it to very much an alpha version. That said, if Microsoft tightens it up, it could be just as good as Firefox 0.9.X. It did have a few things going for it:
1. The tab system is much more visible than Firefox's. I have always believed that if the Mozilla programmers were adverse to leaving tabs mode on by default in Firefox, they should at least have the tabs icon on the bookmarks tool bar so that activating tabs can be easily found. Also, I like some of Microsoft's ideas for handling tab overflow. Its a real problem for someone like me who likes to use up to 50 tabs at a time.
2. The RSS on IE7 Beta is far more robust than Firefox's. Of course Firefox could solve this problem by importing the sage extension into Firefox itself.

Finally, I like that each new major release of Firefox incorporates at least 1 widely used extension. It adds new functions to Firefox while fighting the bloat of a programmer-decided, locked-feature set other browsers have. That said, I hope that extensions are never incorporated for the sake of incorporation and that no more than 3 are ever included at a time.

Posted by: kwerboom at February 13, 2006 8:26 PM

firefox 2 has really cool features.

places (redesigned bookmarks and history)
extended tab browsing (undo close tab)
session saver or atleast crash recovery
visual refresh
anti phishing

just look at Mozilla Wiki

Posted by: milki at February 15, 2006 8:35 AM

Why did not the 'dogfood' land on the 1.8.1 branch as advertized ? in the minutes of the meeting?

Posted by: Missing Places at February 16, 2006 10:20 AM

Despite of the iTunes thing I second Bruce's post that Firefox could do with a bit more innovation.

Posted by: rP2 at February 18, 2006 5:40 AM

Perhaps a Firefox tour would be convenient for new users. This tour should guide users to addons.mozilla.org and present maybe the top 10 extensions. Furthermore edgebreaking features should be outlined, I am thinking of a user tutorial which could do this approriately.

Posted by: rP2 at February 18, 2006 5:48 AM

I am afraid the development team has not enough resources to come up with something really innovative and a 'head turner' to match up with IE7. I am afraid IE7 will give the users enough to convince them to go back to ie7 or to stay with it. I like the idea of scrapbook, but we should really think well what we can do to make FF stand out. The great amount of memory FF uses makes people pissed off and should definitively be tackled, even before 2.0. Is there any place on the web where we can work together with the developers to come up with new ideas, prioritize them, vote, etc.?
I would hate to see IE7 win, my heart really lies with FF!

Posted by: Cheski Frank at February 20, 2006 1:51 AM

it's their own fault if thei already loaded the 1.5 beta, but giving them alpha versions wouldn't be good at all, i think.

Posted by: Slav at February 22, 2006 6:11 AM

Waaaaaa! :-( Keep up the good work and don't rush anything!

Posted by: waaa at February 22, 2006 5:05 PM

You know what would make those NASA shuttles really cool? Forget those boring, under-the-hood safety features. NASA should really be throwing it's money at bionic chimp pilots and soup cozies that keep my soup hot even in the vacuum of space. Unfortunately, I know nothing of soup dynamics in a vacuum, future-chimps or space shuttles. Nor do I work at NASA. However, if I *did* know a thing or two about those things I might head over to NASA and ask for a job in the Wow! department.

Ferdinand: What you describe is not FireFox. It's IE5.5 and Windows Vista with a dash of antitrust thrown in for good measure. What makes FireFox great is that I can use bookmarks and you can use scrapbooks, but we can both use the best standards-based browser available today. The FF development team should be working on *nothing but* under-the-cover changes and leaving the Wow features to the extension developers, then let each of use choose what extensions make FF great for our own personal requirements.

Kudos to the FF dev team for pushing back on an Alpha release until they are good and pleased with the product. Much better to wait than ship and patch-patch-patch.

Posted by: Chris Bloom at February 27, 2006 7:48 AM

Posted by: chy at March 7, 2006 10:07 PM

is it just me or is fernando insane?

Posted by: guy at March 15, 2006 3:18 AM

©1997-2006 Ben Goodger. All Rights Reserved.
Opinions expressed here are my own, and not those of any organization that I may be affiliated with.
Reload icon is © Stephen Horlander;
Firefox logo is by Jon Hicks, and is a trademark of The Mozilla Foundation.
GetFirefox buttons are from rakaz